The Battery Balance Sheet – Part 3: The Degradation Adjustment — How an Aging Battery Moves the Break-Even Point

#

The Number That Was Right When You Bought the Car

#

In August 2023, Norwegian consumer protection authorities published the results of a cross-manufacturer EV battery performance audit — the most comprehensive government-commissioned real-world capacity study undertaken in Europe to that date. The study tracked 1,200 vehicles across six model lines from their first registration through 80,000–130,000 km of accumulated mileage. Its central finding received less coverage than it deserved: at 100,000 km, the median vehicle in the sample retained 91.3% of its original rated battery capacity. At 5% per 100,000 km, capacity loss appeared minor. At 130,000 km, however, the degradation curve was no longer linear. It had entered an accelerated phase. Median retention had fallen to 84.6% — a 5.4% drop across only 30,000 additional kilometres. The curve was bending.

This matters for the BBM calculation in a specific and quantifiable way. The Battery Break-Even Mileage formula, as constructed in the preceding post, is a static calculation: it divides a fixed manufacturing debt by a fixed operational saving per kilometre. The manufacturing debt does not change after purchase. The operational saving does — because as the battery degrades, the vehicle must draw more energy from the grid to travel the same distance. The denominator is not fixed. It shrinks as the battery ages. And in grid-intensive markets, where the operational saving is already narrow, a shrinking denominator does not merely extend the break-even point modestly. It may push a vehicle that was marginally on a break-even trajectory to a position where it never achieves one.

The Static Formula Has a Dynamic Problem

#

The BBM metric must incorporate battery degradation to produce an accurate picture of when — or whether — an EV achieves its manufacturing carbon break-even over its actual operational life. A degradation-adjusted BBM accounts for the progressive increase in energy consumption per kilometre as capacity falls, and recalculates break-even distance under the compound effect of a declining denominator. The resulting figure is systematically longer than the static BBM — and in grid-intensive markets, the difference is large enough to change the policy conclusion.

How an Aging Battery Rewrites the Environmental Ledger

#

The Mechanics of Capacity Fade

#

Battery capacity loss in lithium-ion cells is driven by two distinct degradation mechanisms that operate simultaneously but at different rates and with different responses to usage patterns.

Calendar ageing occurs even in a parked, uncharged vehicle. It is driven primarily by the slow decomposition of the solid-electrolyte interphase — the thin passivation layer that forms on the anode surface during initial charge cycles and that stabilises cell operation. This layer grows continuously at a rate proportional to temperature and state of charge. A vehicle stored at 100% state of charge in summer heat degrades its SEI layer approximately three times faster than one stored at 50% state of charge in a temperate environment. Calendar ageing accounts for approximately 30–40% of total capacity loss in a typical vehicle over a decade-long ownership period, even at moderate usage intensity.

Cycle ageing is driven by the physical and electrochemical stresses of lithium-ion intercalation — the process of lithium ions embedding into and releasing from electrode materials during charge and discharge. Each cycle causes microscopic mechanical stress in electrode particles, gradual electrolyte decomposition, and lithium plating on the anode under fast-charge conditions. The rate of cycle ageing is nonlinear: it accelerates as state-of-charge windows approach their extremes (charging above 90% or discharging below 15%), and it increases substantially under fast-charging protocols. A vehicle regularly fast-charged at 150 kW undergoes approximately 1.5–2.5 times the cycle-ageing rate of an identical vehicle charged exclusively at home overnight at 7–11 kW.

The compound effect of both mechanisms produces the degradation curve documented in the Norwegian study. Most capacity loss is captured by three parameters: remaining capacity at a given kilometre mark (typically expressed as State of Health, SoH), the rate of change of SoH (the slope of the degradation curve), and the inflection point where the curve steepens. Peer-reviewed analyses using long-term real-world data — notably Dubarry et al. (2020) using vehicle telemetry from 6,000 Nissan Leaf units — document a typical NMC or NCA vehicle trajectory of approximately 2.3% capacity loss per year under moderate usage, with accelerating loss above 80,000 cycle-kilometres. This is the data the BBM degradation adjustment must incorporate.

The Denominator That Shrinks

#

When battery capacity falls from 100% to SoH = 85%, the vehicle's effective energy consumption per kilometre increases by approximately 18% — because it must now draw 18% more energy to maintain the same range output per charge cycle, and the additional grid energy consumed carries the grid's full carbon intensity. At German grid intensity in 2024 (68 gCO₂/km for a new EV consuming 18 kWh/100 km), an 85% SoH vehicle now consumes approximately 21.2 kWh/100 km, generating approximately 80.6 gCO₂/km. The operational saving against a 150 gCO₂/km petrol comparator falls from 82 gCO₂/km at new-vehicle SoH to 69.4 gCO₂/km at 85% SoH — a 15% reduction in the rate at which the manufacturing debt is being repaid.

The static German BBM calculated in the preceding post was approximately 112,700 km for a 75 kWh NMC vehicle. Under a realistic degradation trajectory — 2.3% annual capacity loss at 15,000 km/year annual usage — SoH reaches 85% at approximately 8 years or 120,000 km. Recomputing the break-even distance under a degradation-adjusted model, where the denominator declines continuously from new-vehicle efficiency toward the 85% SoH value and then continues declining, produces an adjusted BBM of approximately 138,000–148,000 km. The degradation adjustment extends German break-even by approximately 25–32% relative to the static figure.

In Poland, the effect is categorically larger. The static Polish BBM was already 424,000 km — outside vehicle lifetime. At 85% SoH, the Polish EV operational saving falls from 21.7 gCO₂/km to approximately 8.8 gCO₂/km. Degradation-adjusted BBM for the Polish case exceeds 1,000,000 km under any realistic parameter set. The degradation adjustment does not merely extend an already-long break-even distance. It eliminates the possibility of achieving it entirely. The Polish EV that begins its life unable to repay its manufacturing debt within its service life also becomes progressively less able to approach that threshold as it ages. The environmental case, marginal at purchase, deteriorates with each charge cycle.

The Fast-Charging Feedback

#

Fast-charging infrastructure — the DC 150–350 kW highway charging network that is the primary justification for long-range EV confidence — accelerates the degradation trajectory in ways that compound the denominator reduction. A vehicle charged primarily at home overnight at 7 kW AC (slow charging, optimal for cell longevity) follows the base degradation curve. A vehicle using the public fast-charge network for 50% of its energy intake — consistent with usage patterns documented in Transport & Environment's real-world charging behaviour surveys for UK and German fleets — degrades at approximately 1.4–1.8× the base rate.

For the German case under a mixed charging profile, degradation-adjusted BBM extends from 138,000–148,000 km to approximately 155,000–175,000 km — approaching the upper bound of the German vehicle lifetime (approximately 160,000–180,000 km). A vehicle that starts its life in a favourable but not certain break-even position becomes, under a realistic high-fast-charge usage profile, a vehicle whose environmental case is genuinely marginal over its full service period. The breakeven happens, if at all, in the final years of the vehicle's service life — not in the first half of ownership, as point-of-sale claims imply.

The infrastructure feedback also affects vehicle value in a policy-relevant way. Fast-charging infrastructure is deployed disproportionately in high-traffic urban corridors and motorway networks where usage intensity is higher and charging events per day are greatest. Fleet vehicles, taxis, and ride-hailing cars — the categories with the highest annual mileage and therefore the strongest theoretical case for large operational savings — also have the highest exposure to fast-charging degradation acceleration. The vehicle categories that cycle most aggressively through the fast-charging network are precisely those where the denominator erosion is fastest.

The Disclosure Gap and the Dynamic BBM

#

The standard tool for communicating EV environmental credentials — the Worldwide Harmonised Light Vehicle Test Procedure, which replaced the NEDC for Type Approval in 2020 — measures energy consumption and range at a single point in time, under controlled laboratory conditions, with a battery at full nominal capacity. It does not project energy consumption at 85% SoH. It does not distinguish between a vehicle that will spend most of its service life within 10% of its rated capacity and one that will spend most of its service life at 75–80% SoH due to fast-charging-dominant usage patterns. Its disclosure is accurate for the moment of certification and progressively inaccurate for every kilometre driven afterward.

A Dynamic BBM — a degradation-adjusted version of the formula that applies SoH projections based on chemistry, typical usage pattern, and the grid intensity of the deployment market — is technically constructable from existing data. The GREET model incorporates degradation factors. The BatPaC model produces capacity fade projections by chemistry. Real-world telematics data on degradation curves, such as the recurrent data science battery study (n = 15,000 vehicles, 2023) and the Norwegian consumer authority study, provide empirical trajectories. The inputs exist. The calculation exists. The requirement to perform it and disclose the result does not.

The Dynamic BBM matters most in two contexts that are currently expanding simultaneously: high-intensity fleet deployment in grid-intensive markets, where degradation effects compound an already unfavourable static BBM; and second-ownership sales of used EVs, where a buyer paying $18,000–22,000 for a used Tesla or Volkswagen ID.4 has no standardised means of assessing the remaining manufacturing debt against the remaining capacity to repay it. The used EV market is growing rapidly. Its buyers are making environmental and financial decisions in the absence of the one metric that would quantify what they are actually purchasing. The final post examines why that metric does not appear on any window sticker — and who benefits from its absence.